Nexus : An Agentic Framework for Time Series Forecasting
Summary
Nexus introduces a multi-agent framework that decomposes time series forecasting into specialized stages, integrating numerical patterns and contextual information using LLMs, achieving state-of-the-art results on benchmarks.
View Cached Full Text
Cached at: 05/15/26, 04:23 AM
Paper page - Nexus : An Agentic Framework for Time Series Forecasting
Source: https://huggingface.co/papers/2605.14389
Abstract
Nexus is a multi-agent forecasting framework that decomposes time series prediction into specialized stages, enabling effective integration of numerical patterns and contextual information for improved forecasting accuracy and explainability.
Time series forecasting is not just numerical extrapolation, but often requires reasoning with unstructured contextual data such as news or events. While specializedTime Series Foundation Models(TSFMs) excel at forecasting based on numerical patterns, they remain unaware to real-world textual signals. Conversely, while LLMs are emerging as zero-shot forecasters, their performance remains uneven across domains and contextual grounding. To bridge this gap, we introduce Nexus, amulti-agent forecastingframework that decomposes prediction into specialized stages: isolating macro-level and micro-leveltemporal fluctuations, and integratingcontextual informationwhen available before synthesizing a final forecast. This decomposition enables Nexus to adapt from seasonal signals to volatile, event-driven information without relying on external statistical anchors or monolithic prompting. We show that current-generation LLMs possess substantially stronger intrinsic forecasting ability than previously recognized, depending critically on how numerical and contextual reasoning are organized. Evaluated on data strictly succeeding LLM knowledge cutoffs spanning Zillow real estate metrics and volatile stock market equities, Nexus consistently matches or outperforms state-of-the-art TSFMs and strong LLM baselines. Beyond numerical accuracy, Nexus produces high-qualityreasoning tracesthat explicitly show the fundamental drivers behind each forecast. Our results establish that real-world forecasting is an agentic reasoning problem extending well beyond onlysequence modeling.
View arXiv pageView PDFAdd to collection
Get this paper in your agent:
hf papers read 2605\.14389
Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash
Models citing this paper0
No model linking this paper
Cite arxiv.org/abs/2605.14389 in a model README.md to link it from this page.
Datasets citing this paper0
No dataset linking this paper
Cite arxiv.org/abs/2605.14389 in a dataset README.md to link it from this page.
Spaces citing this paper0
No Space linking this paper
Cite arxiv.org/abs/2605.14389 in a Space README.md to link it from this page.
Collections including this paper0
No Collection including this paper
Add this paper to acollectionto link it from this page.
Similar Articles
Detecting Time Series Anomalies Like an Expert: A Multi-Agent LLM Framework with Specialized Analyzers
The article introduces SAGE, a multi-agent LLM framework for time-series anomaly detection that uses specialized analyzers to improve interpretability and reliability. It demonstrates superior performance over baselines on three benchmarks and enhances diagnostic reporting through structured evidence consolidation.
QuantAgent: Price-Driven Multi-Agent LLMs for High-Frequency Trading
QuantAgent is a multi-agent LLM framework designed specifically for high-frequency trading, using four specialized agents (Indicator, Pattern, Trend, Risk) to make rapid, risk-aware decisions based on short-horizon signals. In zero-shot evaluations across ten financial instruments including Bitcoin and Nasdaq futures, it outperforms existing neural and rule-based baselines in predictive accuracy and cumulative return.
Retrieval Mechanisms Surpass Long-Context Scaling in Time Series Forecasting
This academic paper challenges the effectiveness of long-context scaling in time series forecasting, demonstrating that retrieval-based methods outperform standard architectures like PatchTST and foundation models such as Chronos and Moirai.
TradingAgents: Multi-Agents LLM Financial Trading Framework
This paper introduces TradingAgents, a multi-agent LLM framework that simulates real-world trading firms to improve stock trading performance. It utilizes specialized agents for analysis and risk management, demonstrating superior results in cumulative returns and Sharpe ratio compared to baselines.
TMAS: Scaling Test-Time Compute via Multi-Agent Synergy
TMAS introduces a multi-agent framework that enhances large language model reasoning by scaling test-time compute through structured collaboration and hierarchical memory systems. The approach uses specialized agents, cross-trajectory information flow, and hybrid reward reinforcement learning to improve iterative scaling and stability on challenging reasoning benchmarks.